Awesome-Diffusion-LLM
paper list, tutorial, and nano code snippet for Diffusion Large Language Models.
https://github.com/yczhou001/Awesome-Diffusion-LLM
Last synced: 19 days ago
JSON representation
Application
- Reinforcing the Diffusion Chain of Lateral Thought with Diffusion Language Models - 2025.05-red)
- Table-to-Text Generation with Pretrained Diffusion Models
- Diffusion of Thought: Chain-of-Thought Reasoning in Diffusion Language Models
- DiffusionNER: Boundary Diffusion for Named Entity Recognition
- Fine-grained Text Style Transfer with Diffusion-Based Language Models
Continuous-Time Diffusion Model
- Unifying Continuous and Discrete Text Diffusion with Non-simultaneous Diffusion Processes
- TESS 2: A Large-Scale Generalist Diffusion Language Model - 2025.02-red)
- Difformer: Empowering Diffusion Models on the Embedding Space for Text Generation
- Diffusion Guided Language Modeling
- TESS: Text-to-Text Self-Conditioned Simplex Diffusion
- Transfer Learning for Text Diffusion Models
- Latent Diffusion for Language Generation
- AR-Diffusion: Auto-Regressive Diffusion Model for Text Generation
- Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise
- DiffuSeq-v2: Bridging Discrete and Continuous Text Spaces for Accelerated Seq2Seq Diffusion Models
- DeTiME: Diffusion-Enhanced Topic Modeling using Encoder-decoder based LLM
- How Does Diffusion Influence Pretrained Language Models on Out-of-Distribution Data?
- SSD-LM: Semi-autoregressive Simplex-based Diffusion Language Model for Text Generation and Modular Control
- Glyphdiffusion: Text generation as image generation
- Dinoiser: Diffused conditional sequence learning by manipulating noises
- Diffusion-LM Improves Controllable Text Generation
- Latent Diffusion Energy-Based Model for Interpretable Text Modeling
- DiffuSeq: Sequence to Sequence Text Generation with Diffusion Models
- Continuous diffusion for categorical data
- Seqdiffuseq: Text diffusion with encoder-decoder transformers
Diffusion Language Model for Multimodal Understanding
- LaViDa: A Large Diffusion Language Model for Multimodal Understanding - 2025.05-red)
- MMaDA: Multimodal Large Diffusion Language Models - 2025.05-red)
- LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuning - 2025.05-red)
- Dimple: Discrete Diffusion Multimodal Large Language Model with Parallel Decoding - 2025.05-red)
- Unified Multimodal Discrete Diffusion - 2025.03-red) [ CODE ]
- DiffusionVL: Translating Any Autoregressive Models into Diffusion Vision Language Models - 2025.12-red) [ CODE ]
Discrete Diffusion Model
- Esoteric Language Models - 2025.06-red) [ CODE ]
- Accelerating Diffusion LLMs via Adaptive Parallel Decoding - 2025.06-red)
- DINGO: Constrained Inference for Diffusion LLMs - 2025.05-red)
- Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding - 2025.05-red) [ CODE ]
- LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Models - 2025.05-red)
- Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusion - 2025.05-red)
- Diffusion vs. Autoregressive Language Models: A Text Embedding Perspective - 2025.05-red)
- dKV-Cache: The Cache for Diffusion Language Models - 2025.05-red)
- CtrlDiff: Boosting Large Diffusion Language Models with Dynamic Block Prediction and Controllable Generation - 2025.05-red)
- Insertion Language Models: Sequence Generation with Arbitrary-Position Insertions - 2025.05-red)
- d1: Scaling Reasoning in Diffusion Large Language Models via Reinforcement Learning - 2025.04-red)
- Target Concrete Score Matching: A Holistic Framework for Discrete Diffusion - 2025.04-red)
- Remasking Discrete Diffusion Models with Inference-Time Scaling - 2025.03-red)
- Large Language Diffusion Models - 2025.02-red) code
- Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models - [[💻]] Code
- Beyond Autoregression: Fast LLMs via Self-Distillation Through Time
- Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Data
- Masked Diffusion Models are Secretly Time-Agnostic Masked Models and Exploit Inaccurate Categorical Sampling
- Scaling Diffusion Language Models via Adaptation from Autoregressive Models
- Scaling up Masked Diffusion Models on Text
- Speculative Diffusion Decoding: Accelerating Language Generation through Diffusion
- Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution
- Discrete Flow Matching
- Simple and Effective Masked Diffusion Language Models
- Simplified and Generalized Masked Diffusion for Discrete Data
- A Reparameterized Discrete Diffusion Model for Text Generation
- Diffusion-NAT: Self-Prompting Discrete Diffusion for Non-Autoregressive Text Generation
- Diffusion Glancing Transformer for Parallel Sequence-to-Sequence Learning
- Diffusion Language Models Can Perform Many Tasks with Scaling and Instruction-Finetuning
- DiffusionBERT: Improving Generative Masked Language Models with Diffusion Models
- DiffusER: Discrete Diffusion via Edit-based Reconstruction
- Likelihood-Based Diffusion Language Models
- A Cheaper and Better Diffusion Language Model with Soft-Masked Noise
- A Continuous Time Framework for Discrete Denoising Models
- Structured Denoising Diffusion Models in Discrete State-Spaces
- Argmax Flows and Multinomial Diffusion: Learning Categorical Distributions
Nano code
- GUIDELINES.md - GSAI/LLaDA) for parts of our implementation. Thanks for their valuable contributions!